When the pandemic arrived in early 2020, rheumatology received more attention than many other medical subspecialities. There was a tremendous amount of interest in whether hydroxychloroquine was preventive or ameliorative, along with controversy over whether patients with autoimmune disease or those who were immunocompromised were more or less likely to become infected, and increasing reports of rheumatic syndromes appearing among patients with COVID-19, including coagulopathies, cytokine storm syndromes, multisystem inflammatory syndrome in children and adults, and vasculopathies presenting with features of chilblains or Raynaud's phenomenon. Ultimately, agents used to manage patients with rheumatic diseases, such as tocilizumab, dexamethasone, and baricitinib, were found to be beneficial for pulmonary and other complications of severe COVID-19. More recently, guidance documents have attempted to clarify the role for vaccinations in immune-compromised patients, especially those on anti-inflammatory agents. Rheumatologists' proactive response in developing global research collaborations has created new opportunities. For example, the COVID-19 Global Rheumatology Alliance started a registry that began to enrol patients with rheumatic diseases infected with SARS-CoV-2 within the first weeks of the pandemic.

Ultimately, nearly a year elapsed before two of the most important outstanding questions were meaningfully addressed: whether patients with rheumatic disease are more or less likely to become infected with SARS-CoV-2, and whether the severity of illness in these patients was different to that in the public at large. In other words, after hearing about many cases, we finally have a denominator so comparisons can be made. In The Lancet Rheumatology, David Saadoun and colleagues present a multicentre study of 3028 patients with rheumatoid arthritis, axial spondyloarthritis, systemic lupus erythematosus, Sjogren's syndrome, and giant cell arteritis, from six referral centers in Europe (France, Germany, Italy, Portugal, Spain, and the UK). Patients were derived from well described and well characterized cohorts, and the results are of great interest. The reported prevalence of SARS-CoV-2 antibodies, hospital admissions, and deaths were the same as in the general population. 519 (20·6%) of 2514 patients had treatment changes, of which 125 (24·1%) were due to the pandemic, and a 654 (21·6%) of 3028 patients had at least one disease flare—a proportion that is probably not dissimilar to that seen in normal times. C-reactive protein was found to be a reliable marker of symptomatic COVID-19 (odds ratio 1·18, 95% CI 1·05–1·33; p=0·0063).

However, the study had many limitations. The reported SARS-CoV-2 seropositivity rates were relatively low (eg, several states in the USA had rates that were several times higher), patients in the study were usually only tested on one occasion, and there were too few hospital admissions or fatal events to generate any real conclusions. Racial and ethnic diversity among the patients followed up is also likely to have been limited as established European rheumatology cohorts are overwhelmingly white.

More than 110 000 articles on COVID-19 have appeared in the peer-reviewed literature, but fewer than 1% of them involve patients with autoimmune disorders. As such, the impact of published data from patients with autoimmune diseases largely remains unclear, and many questions remain. Since patients with systemic lupus erythematosus have higher levels of type I interferon, might this minimize the severity of COVID-19? Calabrese and colleagues raised the concern that anti-interferon agents might be less protective of certain infections and are associated with a relative high prevalence of herpesvirus infections. Patients with severe COVID-19 have also been found to harbor 40 times more autoantibodies against type 1 interferon than healthy individuals.